Photovoltaic (PV) solar energy generating capacity has grown by 41 per cent per year since 20091. Energy system projections that mitigate climate change and aid universal energy access show a nearly ten-fold incr. In the International Energy Agency's (IEA) Sustainable Development Scenario, 4,240 GW of. Earth observations analysed with machine learning offer the only feasible method to produce a dataset like this on a global scale. The maturation of computer vision using CNNs12,13. The machine learning pipeline is deployed on the global corpus of Sentinel-2 and SPOT6/7 imagery using Descartes Labs cloud computation infrastructure. Assuming that installations. With our dataset of installation geometries we are able to generate insight into global land-cover patterns of PV solar energy sites. Land use for renewable energy is an urgent area of st. Our global survey of non-residential PV solar energy installations, using machine learning and remote sensing, has generated a public global database of 68,661 spatially l.